Intimate partner violence against women is a widespread phenomenon that is the cause of many deleterious health and social consequences. This study examines the impact of some risk factors on partner violence in the eastern region of Iran, using path analysis. The study used a population-based cross sectional study design. In this study, 251 married women who were referred to the health centers were selected through a proportionally stratified and randomized sampling method. Domestic violence was measured using Conflict Tactics Scale and the socio-demographic variable was assessed by a self-report questionnaire. Bayesian Structural Equation Modeling was used for evaluating the overall path analysis and the direct and indirect p-value was estimated by Bootstrap method. AMOS and SPSS software were used to analyze data. The prevalence of overall violence was 78.1%, with 37.8% and 0.8% of women reporting minor and severe violence, respectively, and 39.8% reporting both severe and minor forms of violence. Psychological violence was the most common type of violence reported (66.5%). The model showed that husbands’ drug abuse and women's higher level of education compared to their husbands were the first and second most important factors that significantly and directly influenced the violence. The women's attitude, however, had the least effect on the violence. The findings indicated that higher educated women and women with addicted husbands were more likely to experience violence. Treating the drug abuse disorders, especially mental disorders, using behavioral couple's therapy, as well as modifying certain traditional and cultural biases against women's empowerment are suggested.